Data was pulled using .sql query from an institutional database to include: the deidentified student, course name, term, letter grade, and GPA for both the "first" course, that is,  the targeted three courses listed in the above chart, and the "second", or subsequent, course. A "third" course was also populated in the case that a student skipped the immediate subsequent term entirely but re-enrolled in coursework the following term. 

​

   Criteria used to populate raw data for analysis:

 

• Courses into which students enrolled the immediate next ("second" term)

• If a student skipped a second term, their subsequent course was identified in the following term ("third" term) 

• Any "fourth" term data was not identified or considered

• If a student took multiple subsequent summer courses and withdrew from all of them, and then enrolled in a third subsequent course, the third subsequent was considered as the subsequent course, since it appeared the student withdrew based on something other than the courses being too difficult

• Required courses that immediately followed at the same or next level (not a lower level course)

• Subsequent courses had to have ten or more students entering so as to avoid the few students who may have taken a course in an isolated part of college and, therefore, may not have been representative of a trend

• Excluded courses were: Courses excluded were AGB-101, AGB-331, BUS-161, GLS-110, and SDV-109

• If students dropped before the refund date (first two weeks of class), they were not included; otherwise, data includes all students past that point

​

When the raw data was populated, it was then processed to eliminate students who had been issued a final grade of W in their "first" course (not the subsequent), as this may have invalidated the data set, since it was not known why or when exactly students stopped attending a course (i.e. early in the semester or right before drop date); Ws ultimately meant they did not finish and were not included in the analysis. 

​

Then, any duplication in the first-subsequent courses for any one student was erased, since some were duplicated if there was a third course. Also, any first-subsequent courses that had no inherent or logical connection referencing language or skill domain was deleted (i.e. it is irrelevant how students in the reading course  did in a subsequent speaking course).  This reduced the data set from approximately 2,000 rows to 550. 

​

Next, all possible "pathways" were identified (see "Pathways" tab in chart, below), labeled using a naming convention, and sorted within the Excel document using a nested if-clause code added to a new column. This allowed specific individual pathways to be displayed for analysis. For example, when a particular pathway is displayed individually, we can more clearly say: I see X number of students who took X course using the X grading structure followed by X course, and here’s what happened. This "pathway" sorting also allowed number grades to be averaged and compared, as well as the compilation of a secondary measure of counting of how many A, A-, B+, B, etc. there were, since numeric averages may not account for range (i.e. all students got a B and some who got all As and Cs still may have the same average of B).

 

The data was then interpreted and reported on the document's "Overall Results" and "Pathways Results" tabs, below.

​

Results Tab Key

• T = Traditional grading structure (points and weighted categories)

• R = Competency-based grading structure, revised version (students must demonstrate competency all outcomes before passing the class)

• Curr= Current course

• Sub = Subsequent course

• Cond= Condensed

• Sts = Students

​

Raw Data Tab Key

• Code = Deidentified student; this column has redundant number values that represent the same students but with multiple data points

• First Term = Semester and year of the original language course that used either a competency-based or traditional grading structure

• First Course = The original language course that used either a competency-based or traditional grading structure

• TRAD/REV = Traditional grading of points or weighted categories (TRAD); competency-based grading, revised version (REV)

• Second Term = Semester and year of the subsequent course

• Second Course = A student's subsequent course from the "first course", whether they enrolled in the immediate subsequent term or skipped it and enrolled in a third term